Gestalt is live. What building it revealed about the problem it solves
-
Moe Hachem - March 27, 2026
I built Gestalt because I kept running into the same wall.
I was building multiple products simultaneously — each one a system with its own architecture, its own decisions, its own dependencies. I needed to hold the map in my head while also executing at the file level. No tool I had let me do both. I could zoom out and lose the detail, or zoom in and lose the map. The two views never coexisted.
So I built the tool I needed.
Gestalt is a visual system mapping tool with fractal navigation. Every node can contain an entire map. You can nest indefinitely — strategy inside architecture inside component inside decision — and navigate between levels without losing your orientation. It took eight days to build from scratch to production. It launched February 4, 2026.
I’m telling you about the build before I tell you about the product because the build is the argument.
The fractal problem
Most tools solve for either overview or detail. Project management tools give you a board. Diagramming tools give you a canvas. Both are flat. They assume you can hold the vertical — the connection between the decision that was made and the component that expresses it — in your head.
For simple systems, you can. For complex ones, you lose threads. Not because you’re not smart enough. Because the tool isn’t structured to hold the thread for you.
The fractal model solves this by making depth a first-class navigation primitive. You don’t switch views. You drill. Every node is a potential universe. You decide how deep any given branch needs to go.
The practical consequence: you can map a system at the level of “what are the five products and how do they relate” and also map a system at the level of “what are the seventeen components inside this one feature and what depends on what” — and those two maps are the same map. They’re just at different zoom levels.
What eight days actually means
Eight days from idea to production-grade, security-audited, deployed product. That’s not a sprint. It’s closer to a proof of concept that passed every real constraint.
What made it possible wasn’t speed. It was upfront documentation. Before a single line of code was written, the system architecture was indexed. Every decision had a home. The AI agents building the product could orient themselves without re-explanation at each session.
This is the part most people misunderstand about AI-assisted development. The speed isn’t a feature of the AI. It’s a feature of the structure you build around it. Without SR-SI — the context indexing methodology I’ve been running for three years — eight days becomes eight weeks. The AI doesn’t remember yesterday’s decisions. The index does.
The freemium model and what it signals
Gestalt is free for up to 6 portals. Pro is $19/month.
The free tier is generous by design. The product needs to be used before it can be understood. No amount of description replaces the experience of building your first nested map and realizing you’ve found a structure for something you’ve been carrying in your head for months.
The pricing isn’t the point of this post. The point is the constraint the pricing reflects: if you can’t demonstrate value before asking for money, you haven’t built the right product yet.
